Project Summary Our long-term goal is to develop immune therapies that target aberrantly expressed proteases in blasts and leukemia stem cells. PR1 peptide (VLQELNVTV) is a peptide derived from the leukemia-associated antigens proteinase 3 (P3) and neutrophil elastase (NE), which is presented on HLA-A2 to PR1-specific cytotoxic T lymphocytes (PR1-CTL). During the last grant period, we showed that PR1 is cross-presented by dendritic cells (DCs) and B cells, and we showed the mechanism required proteasome cleavage exogenous P3 and NE taken up by antigen-presenting cells. In previous years of the SPORE grant, we conducted a Phase I-II PR1 vaccine trial in 66 patients with AML, CML, and MDS, and observed immune responses to PR1 vaccine in 58%. However, objective clinical responses were observed in only 11 (16%) patients, and these were limited to patients with low leukemia burden (<10% blasts). We showed that, although highly cytolytic PR1-CTL that expressed high avidity T cell receptors (TCR-??) for PR1/HLA-A2 increased after PR1 vaccination in some patients, they underwent apoptosis by leukemia that expressed high PR1/HLA-A2 surface antigen, resulting in immune tolerance by deletion of high avidity PR1-CTL. Furthermore, although high avidity PR1-CTL could be isolated from umbilical cord blood (CB) units, they are difficult to expand in sufficient quantity ex vivo to be useful as an adoptive cell therapy, thus limiting their therapeutic potential. Therefore, in a novel alternative treatment approach to target PR1, we produced a TCR-like monoclonal antibody (8F4) against PR1/HLA-A2. We have produced a humanized 8F4 (h8F4) with high affinity (KD=7.8 nM) to PR1/HLA-A2 and we showed that h8F4 eliminated AML and leukemia stem cells but not normal human hematopoietic stem cells in preclinical models. With an agreement from industry that supported manufacturing of h8F4, we have produced sufficient clinical grade h8F4, showed it mediated ADCC and apoptosis of AML and LSC in vitro and in vivo, and developed companion assays for PK, anti-idiotype and anti-drug antibody testing. Moreover, we have established PDX models of AML for parallel studies to support a clinical trial. Thus, the goals of this proposal are to (1) translate the discovery of this novel h8F4 monoclonal antibody to the clinic in a first-in-human phase I trial in AML; (2) to determine pharmacokinetics, toxicity, and mode of action; and to characterize the mechanism of action, potential resistance mechanisms and to test novel strategies with an h8F4-based bispecific antibody and an h8F4 chimeric antigen receptor (CAR) T cells to increase the potency of 8F4 to overcome potential treatment resistance.